An important piece of information came out this morning, three previously unreported deaths in Santa Clara County very early on. The earliest of these was on February 6th, more than three weeks before the previously known first death in the US. This is important.
The earliest known case in California had been January 26th, 11 days earlier, in Orange county. Considering that the average time from infection to death is 18 days, and the fact that this case is in a different region of California, this is a pretty clear indication that the virus was spreading in California earlier than previously known. It appears that the case was from community transmission as the victim did not have a history of travel. So it seems likely that there was community transmission in California by around the 20th of January when the first US case was identified.. Now add in the fact that only a small percentage of people who are infected go on to die as a result and it’s not hard to imagine the virus spreading in the US in early January, before the virus was known to have left China. In short, this move the timeline of the virus in the US up by 2-4 weeks.
Why is that important? There are several ways to think about it. We’ve known all along that reported cases and deaths were only a fraction of the real cases and deaths. The longer the virus circulated before detection, the greater the fraction of unreported cases and death is likely to be. This gives some support to the idea that there may be MANY more cases than we realize with the possible result that the virus has a lower fatality rate than we thought. At the same time, though, it shows how we may have missed many deaths. Not only may many of the deaths from “flu-like” symptoms be from COVID, but given that we’ve learned that the virus attacks other organs, we may find that even more of the “surplus” deaths we’ve seen this year have their basis in COVID.
Think about the changes we’ve seen in the last 2-4 weeks. In the last 2 week, our case numbers have roughly doubled and our death numbers have tripled. They were growing even faster before that. In the last month we’ve seen a levelling off of case numbers. I’ve raised the question before whether this levelling off was due to social distancing or limited testing. As you can see in the graph below, the levelling off of cases roughly corresponds to the levelling off of testing. If the virus was circulating in the US much earlier than known, then the role of limited testing in the apparent levelling is likely greater.
The main things to learn from this new information is that we really know very little about the spread of the virus in the US. Really, the only way to fix that is through greatly increased testing, which, as the graph shows, just isn’t happening. We’ll eventually figure some things out through historical analysis, but that will come far too late for us to do anything about it. So, in addition to social distancing, the most effective thing you can do is lobby for increased testing. Without that, we’re just playing blind man’s bluff with the virus, and we’re the ones with the blindfold.
So, numbers. Over the last 5 days, we’ve added an average of about 28,000 cases per day while deaths have increased by about 5% per day. If those rates were to continue for the next 10 days, here’s what it would look like.
| Cases | Deaths | |
| Current | 827,358 | 45,518 |
| 4/22 | 855,586 | 47,976 |
| 4/23 | 883,814 | 50,567 |
| 4/24 | 912,042 | 53,297 |
| 4/25 | 940,270 | 56,175 |
| 4/26 | 968,498 | 59,209 |
| 4/27 | 996,726 | 62,406 |
| 4/28 | 1,024,954 | 65,776 |
| 4/29 | 1,053,182 | 69,328 |
| 4/30 | 1,081,410 | 73,072 |
| 5/1 | 1,109,638 | 77,017 |
We will likely be up over a million cases within the next week.
We will begin to see some states relaxing restrictions. Some, I’m looking at you Georgia, will do so dangerously. We may or may not see reported cases rise much in those states due to limited testing. The things to keep an eye on are deaths and hospitalizations. People who end up in the hospital tend to get tested more frequently than those who don’t. But keep in mind that these numbers will lag the reopening by 2-3 weeks. If we just dropped all social distancing today, it would take a minimum of a week to see much effect, and that would be only if we had easily available testing. Hospitalization numbers would change noticeably in another week, and deaths the week after that. So as we watch some places reopen, remember that the first week of numbers don’t mean anything about whether the reopening was successful or not.
Ok, that’s it for today. Be safe, be healthy, and be happy.